This repository presents an intelligent system that integrates climate forecasting and crop recommendation models to suggest future crops in response to the threat of global warming and climate change on agriculture. The system uses a weighted classifier with the Relative Model Accuracy Equation to improve classification models, achieving 99.8% for each performance metric (precision, recall, and f1 score) on the test set. The main contribution is an intelligent system that uses supervised machine learning to predict the best crop for a specific soil in a particular state of Cuba among 22 possible crops for a given year. The system is built as a python module and released under the MIT open-source license for integration into future software solutions.
The forecasting of climate variables is currently performed using linear regression. To enhance the performance and reliability of the system, the development and evaluation of time series forecasting methods will be undertaken.